You are here

| Cancer Stem Cells

Understanding LGR5 in Colon CSCs



"LGR5-Positive Colon Cancer Stem Cells Interconvert with Drug-Resistant LGR5-Negative cells and are Capable of Tumor Reconstitution"

The existence of a cancer stem cell (CSC) in tumours has been explored thoroughly in recent years, and while evidence mounts in favour of its existence, the debate is far from over (Magee et al., Nguyen et al. and Visvader and Lindeman) although several recent papers have suggested that CSCs do exist in solid tumours in mice (Chen et al., Driessens et al. and Schepers et al.). CSCs in the colon are suggested to be marked specifically by LGR5 (Vermeulen et al.), a Wnt target gene, however the heterogenous nature of normal colon stem cells (Takeda et al.) and CSCs in general (Dieter et al., Gupta et al. and Sharma et al.) have made the definitive identification of cells difficult. Now in a paper published in Stem Cells, researchers from the laboratory of Tatsumi Yamazaki who is associated with Forerunner Pharma Research Co. Ltd. and Chugai Pharmaceutical Co. Ltd. have reported the establishment of human colon cancer cell lines that express LGR5 and possess CSC properties, aided by the generation of LGR5 specific antibodies. Additionally, they show that CSC can interconvert between a proliferative LGR5+ state and a drug-resistant LGR5- state and identify an anti-epiregulin antibody which could mediate a reduction in metastatic disease (Kobayashi and Yamada-Okabe et al.).

The authors initially focused on the generation of LGR5-specific antibodies and the generation of human colon cancer cell lines with CSC properties. Two monoclonal antibodies were generated, 2L36 and 2U2E-2, which specifically recognised LGR5 but not LGR4 or LGR6 following immunostaining and flow cytometric analysis. Additionally, the 2U2E-2 antibody reacted specifically with crypt base columnar cells in the normal human intestine and a good correlation between mRNA expression and cell surface staining for LGR5 was observed in human colon cancer cell lines. Using NOG ((NOD/Shi-scid/IL-2Rγnull) mice (Fujii et al.), human colon cancer xenografts were generated by grafting cells from two tumours, which had clear evidence of epithelial ducts and small budding clusters, into the flank of recipient mice which allowed for the generation of two colon CSC lines (PLR59 and PLR123). Both these lines were heterozygous for the mutant K-Ras while PLR123 carried mutant p53 on one allele and were chosen specifically for their fast growth and their ability to reconstitute tumours with ducts and clusters over 10 passages. LGR5 analysis found few positive cells in the original tumour tissue; 0.01% and 0.04% for PLR59 and PLR123 respectively, however the number of LGR5+ cells in the xenografts increased during passaging, while the tumour initiating activity, which is a generally associated with CSC function and number, also increased from an estimated 0.1% at 5 passages to 0.4% by passage 14.

The CSC cell lines grew poorly in spheroid assays, exhibiting few LGR5+ cells, but grew well in adherent cultures with a population doubling time of 2.5 days with cells having an epithelial morphology. Cells taken from spheroids also had a lower tumour initiating activity as compared to cells from adherent cultures. Adherent cells expressed markers known to be associated with CSCs (LGR5+, ALDH+, CD133+, CD44+, EpCAM+, CD166+, CD24+, CD26+ and CD29+), a pattern which was unaltered after one month in culture. Cells also divided symmetrically under self-renewing conditions, but when grown on Matrigel with foetal bovine serum, divided asymmetrically.

Next, the ability of LGR5+ and LGR5− cells to form colonies in vitro and tumours in vivo was analysed by sorting primary cells from xenografts generated by the inoculation of the LGR5+ cells. LGR5+ cells formed colonies on Matrigel and formed large tumours in NOG mice by 34 days; while LGR5- were unable to efficiently form colonies and gave rise to very small tumours. However, both LGR5+ and LGR5- cells were positive for CD133 and CD166, suggesting that the LGR5+CD133+CD166+ cells represent a specific subpopulation. Further characterisation of LGR5- cells found that LGR5-CD133+ cells could from colonies on Matrigel (1.6% efficiency as compared to 4.3% efficiency for LGR5+CD133+ cells), whereas LGR5-CD133- cells died after seeding.

Drug resistance, believed to be a characteristic of CSCs (Buczacki et al.), was then assessed in LGR5+ cells treated with the anti-cancer drug irinotecan for three days. Irinotecan is a widely used drug, mainly used for colon-cancer in combination with various other chemotherapy agents. Cells stopped proliferating and 50% of cells survived, of which all became LGR5- but retained other colon CSC markers (CD133, CD44, CD166, and EPCAM) and were designated drug-resistant LGR5- cells. Tumour initiating activity in these cells was also low. Interestingly, removal of the drug and replating cells led to the re-expression of LGR5 and the resumption of proliferation, confirmed by Ki67 staining, which coincided with LGR5+ cell staining. DNA microarray analysis determined membrane protein encoding genes which were most representative of the drug-resistant LGR5− cells and demonstrated that the MHC class II-related gene HLA-DMA was very specific. Additionally, the gene Epiregulin (EREG) was found to be expressed in both LGR5+ and drug-resistant LGR5− cells, and using this knowledge LGR5- cells could be identified as HLA-DMA+EREG+ cells. Injection of LGR5- cells into NOG mice led to the weak expression of LRG5 alongside robust HLA-DMA and ERG expression, followed by the appearance of LGR5+EREG+HLA-DMA- cells suggesting that reconstitution of the epithelial tumor hierarchy from LGR5- cells occurs through a transition of cells to a LGR5+ state. ­In vivo analysis of drug-resistant LGR5- cells found that after irinotecan treatment of mice bearing tumours derived from injected LRG5+ cells, LGR5+ cells decreased while there was a significant increase in LGR5-HLA-DMA+ cells, while EREG was observed in both populations of cells. As observed in vitro, removal of the drug allowed for the appearance of LGR5+Ki67+ cells, while LGR5-HLA-DMA+ cells were Ki67-, suggesting that tumour reconstitution occurred through LGR5+ cells.

The specific contribution of EREG was then analysed owing to its expression in LGR5+ cells and drug-resistant LGR5- cells and its general lack of expression in differentiated tumor cells and normal tissues. To this end, an EREG-specific antibody was administered after irinotecan treatment of SCID mice carrying tumours derived from injected LGR5+ cells which delayed tumour growth when administered 4 and 11 days after irinotecan treatment. For metastasized tumours, in which metastatic cells were observed to be EREG+, administration of the EREG-antibody 3 days after tumour injection led to a significant decrease in metastatic potential and tumour size as compared to control non-injected mice. Finally, to corroborate the relevance of this study to the human system, LGR5-/+ cells were studied in human colon cancers. Both LGR5+ and LGR5- cells that were HLA-DMA+EREG+ were found, although these were rare but clinical specimens did contain LGR5+Ki67+ cells and LGR5-HLA-DMA+Ki67- cells.

Overall, this paper has managed to demonstrate the establishment of a CSC colon cancer cell line, the existence of the interconversion of LGR5 positive and negative cells and their importance in tumour survival and re-initiation, and that EREG is a possible target which may be used for the elimination of CSCs. Additionally, it adds significant weight to the existence and importance of CSCs in tumours, as has been suggested in several other recent studies (Chen et al., Driessens et al. and Schepers et al.).



Buczacki S, Davies RJ, Winton DJ.
Stem cells, quiescence and rectal carcinoma: An unexplored relationship and potential therapeutic target.
Br J Cancer 2011; 105: 1253–1259.

Chen J et al.
A restricted cell population propagates glioblastoma growth after chemotherapy.
Nature 2012;23:522-6.

Dieter SM et al.
Distinct types of tumor-initiating cells form human colon cancer tumors and metastases.
Cell Stem Cell 2011;9:357–365.

Driessens G et al.
Defining the mode of tumour growth by clonal analysis.
Nature 2012: 488: 527-530

Fujii E et al.
Establishment and characterization of in vivo human tumor models in the NOD/SCID/gamma(c) (null) mouse.
Pathol Int 2008; 58: 559–567.

Gupta PB et al.
Stochastic state transitions give rise to phenotypic equilibrium in populations of cancer cells.
Cell 2011;146:633–644.

Magee JA et al.
Cancer stem cells: Impact, heterogeneity, and uncertainty.
Cancer Cell 2012;21:283–296.

Nguyen LV et al.
Cancer stem cells: An evolving concept.
Nat Rev Cancer 2012;12:133–143.

Schepers AG et al.
Lineage Tracing Reveals Lgr5+ Stem Cell Activity in Mouse Intestinal Adenomas.
Science 2012: 337: 730-735).

Sharma SV at al.
A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations.
Cell 2010;141:69–80.

Takeda N et al.
Interconversion between intestinal stem cell populations in distinct niches.
Science 2011;334: 1420–1424.

Vermeulen L, et al.
Wnt activity defines colon cancer stem cells and is regulated by the microenvironment.
Nat Cell Biol 2010;12:468–476.

Visvader JE, Lindeman GJ.
Cancer stem cells: Current status and evolving complexities.
Cell Stem Cell 2012;10:717–728.



Study originally appeared in Stem Cells.

Stem CellCorrespondent Stuart P Atkinson reports on those studies appearing in current journals that are destined to make an impact on stem cell research and clinical studies.